Drive real Android & iOS devices and web browsers from natural language for mobile + web QA. 290+ tools across device control, app management, automation sessions, browser automation, and flow recording / replay. Bearer-auth — get a token at robotactions.com → Profile → API Tokens.
The last two health checks of this connector failed. If this server requires authentication, missing or expired test credentials may be the cause. A test profile lets Glama authenticate for health checks and discover tools; it is separate from your personal connections.
If you are the author, claim ownership, then add or update a test profile under Admin → Test Profile.
Scored across 403 tools
Descriptions are unusually detailed and explicitly cross-reference siblings (e.g. native vs web mock scopes, the input="hid" caveat), which rescues many near-duplicates. But the set is riddled with parallel families — four mock stacks (android_devtools_mock_*, android_traffic_mock_*, ios_safari_mock_*, webpage_mock_*), four screenshot tools per platform, tap vs tap_by_text vs uhid taps, and evaluate tools across web_/session_/android_devtools_/ios_safari_ — so an agent must read carefully to pick correctly. Overlap is real and widespread, not incidental.
Nearly everything follows snake_case with a predictable domain prefix (device_, android_devtools_, ios_, ios_safari_, webpage_, web_, session_, flow_recording_, testrail_, azdo_, insights_), making the target platform/scope readable from the name alone. Minor deviations exist in the visual-review cluster (list_visual_reviews, get_visual_review, visual_compare, record_visual_verdict, submit_tool_feedback) where the verb_noun pattern loosens.
403 tools is an extreme mismatch for any coherent tool surface, far beyond the 50+ threshold for score 1. Even accounting for the cross-platform and cross-domain breadth, the count guarantees heavy token cost and misselection risk; many tools are thin variants that could be parameters of a smaller set.
The domain — remote Android/iOS device control and mobile test automation — is covered comprehensively: input, files, network capture and mocks, crashes, performance, accessibility, visual diff, flow recording/replay, TMS integration, and dashboards. Gaps are minor (e.g. session_back has no session_forward counterpart, and the Jira/AzDO/TestRail surfaces are mostly read-and-create with limited update/delete lifecycle).
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
ios_tap_by_label2 fields changedInput schema / properties / inputAdded value: +{
+ "description": "Input path. 'auto' (default) tries the direct input path first — the one that needs no automation session — and falls back to the session only when the direct path is not usable right now (the response says which one ran). On a device set to direct-input-only, an unusable direct path is reported as an error instead of silently switching to the session. 'hid' forces the direct path with no fallback, reporting 'no HID available' on iOS 17.x or with no tunnel. 'session' forces the automation session even when the direct path would have worked.",
+ "enum": [
+ "auto",
+ "hid",
+ "session"
+ ],
+ "type": "string"
+}Input schema / properties / nearAdded value: +{
+ "description": "Direct-input path only: text of a neighbouring element, to pick between several look-alike labels.",
+ "type": "string"
+}ios_find_element1 field changedInput schema / properties / inputAdded value: +{
+ "description": "Lookup path. 'auto' (default) follows the device's mode: direct-input-only devices look up without a session (no coordinates returned). 'session' forces the automation-session lookup, which returns coordinates. 'hid' forces the session-free lookup.",
+ "enum": [
+ "auto",
+ "hid",
+ "session"
+ ],
+ "type": "string"
+}ios_scroll_to_element1 field changedInput schema / properties / inputAdded value: +{
+ "description": "Scroll path. 'auto' (default) follows the device's mode: direct-input-only devices scroll without a session (no coordinates returned). 'session' forces the automation-session scroll, which returns coordinates. 'hid' forces the session-free scroll.",
+ "enum": [
+ "auto",
+ "hid",
+ "session"
+ ],
+ "type": "string"
+}flow_recording_save3 fields changedInput schema / properties / tmsLink / properties / externalId / descriptionPrevious value: -"Case id as the TMS shows it: 'C1234', '4521', 'PROJ-12'"New value: +"Case id as the TMS shows it: 'C1234', '4521', 'PROJ-12', local 'LC-12'"Input schema / properties / tmsLink / properties / provider / descriptionPrevious value: -"Which TMS the case lives in"New value: +"Which TMS the case lives in. 'local' = a case saved in Test Manager (LC-12) for a user with no TestRail/ADO yet"Input schema / properties / tmsLink / properties / provider / enumPrevious value: -[
- "testrail",
- "ado",
- "jira",
- "xray"
-]New value: +[
+ "testrail",
+ "ado",
+ "jira",
+ "xray",
+ "local"
+]flow_recording_save2 fields changedInput schema / properties / tmsLink / descriptionPrevious value: -"The TMS test case this flow automates. Required when the workspace enforces case links."New value: +"The TMS test case this flow automates. REQUIRED when the user has a TMS connected (tests only; fragments never need it)."Input schema / properties / tmsLink / properties / refAdded value: +{
+ "additionalProperties": false,
+ "description": "Where the case lives in the TMS. The server verifies and completes this.",
+ "properties": {
+ "containerId": {
+ "description": "Folder id as the TMS browse tools return it: TestRail '<suiteId>' or '<suiteId>:<sectionId>', ADO '<planId>:<suiteId>' (REQUIRED for ADO)",
+ "type": "string"
+ },
+ "internalId": {
+ "description": "Jira/Xray numeric issue id, if known",
+ "type": "string"
+ },
+ "projectId": {
+ "description": "TestRail project id, ADO project, or Jira project key",
+ "type": "string"
+ }
+ },
+ "type": "object"
+}tms_gap_reportflow_recording_save1 field changedInput schema / properties / tmsLinkAdded value: +{
+ "additionalProperties": false,
+ "description": "The TMS test case this flow automates. Required when the workspace enforces case links.",
+ "properties": {
+ "externalId": {
+ "description": "Case id as the TMS shows it: 'C1234', '4521', 'PROJ-12'",
+ "type": "string"
+ },
+ "provider": {
+ "description": "Which TMS the case lives in",
+ "enum": [
+ "testrail",
+ "ado",
+ "jira",
+ "xray"
+ ],
+ "type": "string"
+ },
+ "title": {
+ "description": "Case title (cached for display)",
+ "type": "string"
+ },
+ "url": {
+ "description": "Link to the case in the TMS",
+ "type": "string"
+ }
+ },
+ "required": [
+ "provider",
+ "externalId"
+ ],
+ "type": "object"
+}insights_board_add_chartinsights_board_apply_changesinsights_board_createinsights_board_getinsights_board_save_templateinsights_board_undoinsights_azdo_browseinsights_azdo_sprint_summaryinsights_azdo_test_summaryinsights_testrail_browseinsights_testrail_summaryinsights_azdo_browseinsights_azdo_sprint_summaryinsights_azdo_test_summaryinsights_testrail_browseinsights_testrail_summaryinsights_azdo_browseinsights_azdo_sprint_summaryinsights_azdo_test_summaryinsights_testrail_browseinsights_testrail_summaryios_fps1 field changedInput schema / properties / durationMs / descriptionPrevious value: -"Sampling window in milliseconds (default 2000)"New value: +"Sampling window in milliseconds (default 3000)"record_visual_verdictjira_designs_probejira_designs_probevisual_tolerance_resetDrive cloud Android phones on US mobile IPs. Read the screen, tap, type, install apps and run tasks.
Drive real devices from your AI Coding tool. Embed a client SDK (Unity, Godot, Flutter, iOS/macOS, Android, React Native, Web) in your app, then capture screenshots, traverse the UI tree, inject taps and key events, and run automated test tasks on the physical device over a secure relay.
Let AI agents control Android and iOS devices on Mobilerun Cloud: screen, input, apps, workflows.
Give your AI agent a real cloud Android phone: push APKs, drive it, get screenshots and logs.
Add one secure layer between your agents and this server.
